Wilfried Mestdagh wrote: > > Hi Arjen, > > AM> No, they will probably not all do the same internally: the result > AM> may be the same, but the "virtual machine" that runs the SQL statements > AM> will very probably do different things - this depends on optimisation > AM> and so on. But for most of us it is mainly the result that counts > AM> (and the time it takes for that result) > > But what should then be the best one in speed ? >
As others have replied, the version(s) where you use count() return(s) only one record. And that can be seen from the sQL statement all by itself. I do not think things can be faster than that: only one record that will actually be returned to the calling program. Regards, Arjen
